記住我
RESEARCH ARTICLE
Year : 2021 | Volume
: 58
| Issue : 4 | Page : 289-296
Evaluation of a new multi-epitope sequence of eight known Leishmania infantum antigens for HVL diagnosis by ELISA and Western blot
Marzieh Taherzadeh1, Moradali Fouladvand2, Bahram Kazemi3
1 Islamic Azad University of Shiraz, Shiraz; Department of Microbiology and Parasitology, Persian Gulf Tropical and Infectious Diseases Research Center, Bushehr University of Medical Sciences, Bushehr, Iran
2 Department of Microbiology and Parasitology, Persian Gulf Tropical and Infectious Diseases Research Center, Bushehr University of Medical Sciences, Bushehr, Iran
3 Cellular and Molecular Biology Research Center, Shahid beheshti University of Medical Sciences, Tehran, Iran
Correspondence Address:
Prof Moradali Fouladvand
Rishehr St., Bushehr University of Medical Sciences, 7514633341
Iran
Source of Support: None, Conflict of Interest: None
CheckDOI: 10.4103/0972-9062.318310
Background & objectives: Leishmaniasis, known as a disease with high prevalence proportion throughout the world, is caused by protozoan parasites. Visceral leishmaniasis is the most severe form of this condition reported sporadically from all regions in Iran. Between different diagnostic tests, serodiagnosis of this infection is of utmost importance in both humans and dogs. Although rK39 ELISA test has been extensively validated in endemic areas, there are currently challenges regarding a more appropriate serodiagnostic test.
Methods: A novel multi-epitope construct was designed consisting of highexposedB cell epitopes using eight important antigens of Leishmania infantum (Gp63, KMP-11, HSP70, CPA, H2A, H3, LACK and TRYP). Our artificial sequence, a Multi-epitope Recombinant Protein (MRP), was consequently produced and purified. Then, immunoreactivity was investigated by ELISA test and western blotting as well.
Results: In the present study, the cutoff value (1.052) for the new MRP-ELISA was determined by receiver operator characteristic (ROC) curve analysis using 35 known positive and 20 known negative HVL sera previously tested for antibodies to L. infantum by DAT, showing a sensitivity of 93.1% and a specificity of 77.4%. The blotting test also showed a favorable band to detect visceral leishmaniasis.
Interpretation & conclusion: According to the results, this new antigen had acceptable potential in detecting VL positive cases once western blotting was utilized, but the ELISA test did not proceed as expected for detecting true negative cases, probably due to some optimization issues.The present study is a promising start.
Keywords: Multi-epitope Recombinant Protein (MRP); Leishmania infantum; Chimeric Antigen; ELISA; Immunological Assay; HVL
Leishmaniasis is a parasitic protozoan disease caused by intracellular parasite belonging to the genus Leishmania. The clinical manifestations of the disease range from self-limiting cutaneous lesions to destructive mucosal diseases as well as sub-clinical to fatal visceral infections. Visceral leishmaniasis (VL) as a life-threatening vector-borne disease is a prevalent health problemif it is not properly diagnosed and treated[1]. The disease is endemic in several parts of Iran including Ardabil, East Azarbayjan, Fars, Bushehr, Kerman and Qom and North Khorasan provinces.99.0% of VL cases have been found among children up to 12 years old of these regions. Leishmania infantum (L. infantum) is the main causative agent of human as well as canine VL in Iran[2],[3].
The genome of the Old World species of Leishmania can be organized into 36 heterologous chromosomes with 8000 genes[4]. Among known genes, Cysteine Protease-A (CPA) and kinetoplastid membrane protein-11 (KMP-11) have been found to be highly immunogenic[5]. KMP-11 able to elicit a relatively strong antibodies response in the natural course of leishmaniasis and Immunoglobulin G1 (IgG1) is the major antibody subclass in this pathway[6]. Cysteine proteases (CPs) are virulence factors essential in Leishmania parasites with key roles in many protozoal and viral agents. The bulk of parasitic CPs, characterized by potential immunogenic properties, can also meet diagnostic purposes and vaccine targets[7]. Among indicators, leishmanolysin (GP63) is a surface virulence factor from protease group. This antigen plays an important role in protozoan survival within host macrophages[8]. Some in-tracellular proteins have been similarly described as immunodominant antigens in the course of Leishmaniasis such as heat shock protein 70 (HSP70) which is regarded as a major target of the humoral immune response[9]. Moreover, 78% and 81% of the sera from dogs contained significant levels of antibodies against histones H3 and H2A; respectively[10],[11]. For the first time in the last decade, it was also discovered that dogs infected with L. infantum had developed humoral immunity against try-paredoxin peroxidase(TRYP) and Leishmania-activated C-kinase (LACK) antigens[12]. Many studies had further demonstrated that these antigens could be considered as important immunogens[13],[14].
An accurate diagnosis is thus required to properly manage VL and to prevent dissemination of the disease. So, highly sensitive and specific diagnostic tests are needed for an early and correct diagnosis[15]. Sensitivity rates of different parasitological diagnostic tests for VL also vary from 50% to 99% based on the sample type[16],[17]. Given the difficulty in obtaining a proper sample and the need for expertise and equipment for parasitological diagnosis of VL and molecular testing costs, serological methods are preferable . So far, various methods have been introduced in this line. Firstly, crude or soluble antigens of promastigotes or amastigotes are employed in enzyme-linked immunosorbent assay (ELISA) test; however, their use is not prioritized because of cross-reactivity as a common challenge in these antigens[16],[18]. The second priority to VL diagnosis is recombinant antigens that have been recently used[19],[20]. Accordingly, various serological methods characterized by varying sensitivity and specificity are available for VL diagnosis[14],[21]. The common serodiagnostic methods such as immunochromatographic (ICT-rK39) strip test and Direct Agglutination Test (DAT) that are being widely used in endemic countries have also their own drawbacks like false positive[22]. In Spain, rK39-ICT and DAT have exhibited acceptable sensitivity and specificity but combination with other tests is required for highly sensitive diagnosis of VL cases[23]. Lower sensitivity of ICT-rK39 has limited its use for VL diagnosis in HIV-coinfected patients. DAT is a serological test with good reproducibility, sensitivity and specificity in Iranian settings butit need prolonged incubation times (12 to 18 h) and serial serum dilutions[16]. Maybe, one of the disadvantages of serological tests is that the antigen used does not have enough diversity of antigenic determinants to respond to an extensive range of antibodies from patient serums. Accordingly, one of the main objectives of the present study was to produce recombinant protein having more antigenic diversity than that in available tests. In this study, a novel synthetic gene comprised of different key epitopic antigens of L. infantum was designed and expressed in the prokaryotic host. Then, validity and suitabilityof this antigen was investigated in Western blot and ELISA tests for detecting human VL positive serums.
Material & MethodsPrimary bioinformatics investigation
To select the epitopes of L. infantum, the Immuno Epitope Database (IEDB) was used and the best experimentally tested linear B-cell epitopes with high scores (GP63, CPA, HSP70, KMP-11, H3, and H2a) of L. infantum (strain JPCM5) were chosen based on the validated data in the webserver. LACK and TRYP antigens had no data in the IEDB thus the prediction programs in the server were utilized to pick out the linear B-cell epitopes (www.tools. iedb.org/bcell). Among the introduced epitopes, only one was selected and employed as a duplicate in different positions of the synthetic gen construct.
Multi-epitope sequence
The selected epitopes were linked together using a short flexible glycine-serine linker (4G1S) to produce the construct[24]. The single synthetic sequence encoding all the selected epitopes was also reverse translated into the amino acid sequence and 3D structure were checked by I-TASSER server [Figure 1]. Besides, the construct was tagged via six-histidine-coding sequence in order to simplify the purification of the recombinant protein followed by BamHI restriction site and two stop codons (i.e. TAA and TAG) at the C-terminus (3′ end). On the other hand, a start codon and an Ncol restriction site were placed at the N-terminus of the construct (5′ end). In order to maximize protein expression in the prokaryotic host (i.e. Escherichia coli: E. coli), codon optimization of the construct sequence was carried out by JCat. The amino acid sequence of the MRP is as follows (Note: the flexible linkers are as bold and underlined)
Figure 1: The tertiary structure of MRP. The 3D structure predicted by I-TASSER software. The structure is shown in ribbon model.
Cloning, expression, and purification of multi-epitope sequence
The construct was synthesized and cloned into plasmid pET-25b by Biomatik’s company (USA), and then sub-cloned in expression vector pET-32a. Generally, E. coli Bl21 (DE3) cells were transformed with the pET-32a containing a multi-epitope construct using Calcium Chloride procedure. The bacteria were also grown in a terrific broth containing 100μg/ml ampicillin at 37°C until OD600=0.5. Immediately, the isopropyl β-D-1-thiogalactopyranoside (IPTG) with a final concentration of 0.5mM was added to induce gene expression, the bacterial culture was shacked overnight at 18°C (Novagen I 2005; pET system manual), and then the lysate was checked by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) technique and confirmed via Western blotting[25].
The recombinant protein was also purified through Protino® Ni-NTA Agarose in accordance with the manufacturer’s instructions under native conditions (www.mn-net.com). The identity and the purity of the eluted MRP were additionally confirmed using Western blotting with anti-6x-His tag monoclonal (Invitrogen, USA). The purified MRPwas finally quantified via the bicinchoninic acid (BCA) assay and stored at -20°C until its use.
Serum sampling
A total of 35 human VL serums were received from two important endemic regions of Iran including 22 and 13 serums from the cities of Tabriz and Shiraz; respectively. Also, 20 negative serum samples (that is, 10 sera from volunteer healthy individuals at Bushehr University of Medical Sciences (BUMS)and 10 sera previously tested for antibodies to VL by DAT from sick people of endemic region) and 10 serum samples of individuals affected with other infectious diseases (i.e. 4 Toxoplasma positive sera, 2 malaria positive sera, 2 cutaneous leishmaniasis positive sera, and 2 hydatid cyst positive sera) were also prepared and tested in terms of their cross-reactivity. The VL positive sera were checked again by the IFA test in parasitological laboratory of BUMS [Figure 2].
ELISA test
Initial optimization of the ELISA test was performed based on different concentrations of MRP and primary and secondary antibodies. All the samples were tested by ELISA using the standardized principle[26].The test was also performed in flat bottom 96-well microplates (Nunc- MaxiSorp™, USA) as follows: ELISA plates were coated with 0.5, 1, 1.5 and 2μg/ml of purified MRP and incubated at 4°C overnight. Serum samples at 1:200, 1:500, 1:1000, and 1:2000 dilutions and peroxidase-conjugated goat anti-human IgG at 1:10000, 1:50000 and 1:100000 dilutions were further checked. The enzymatic activity was detected through the TMB substrate (BioLegend). The optical density was finally measured at 490 nm (OD 490) using an ELISA plate reader. It should be noted that all the tests were performed in triplicate.
Statistical analysis
MRP sensitivity and specificity was calculated using DAT and IFAresults as comparative tests. The receiver operating characteristic (ROC) curve with 95% confidence interval was created byGraphPad Prism (Version 5).The software allowed for the estimation of the best specificity and sensitivity for the use of the serological method as a diagnostic test. The cut-off value for the ELISA test was furthermore defined as the optimum scale of sensitivity and specificity[27].
ResultsThe selected epitopic antigens and the related amino acid sequences of L. infantum were utilized in the multiepitope construct as presented in [Table 1] and [Table 2].
Table 1: Selected amino acid sequences of five experimentally tested epitopes of L. infantumConfirmation of multi-epitope construct
Multi-epitope sequence was synthesized and cloned between Ncol and BamHI restriction sites of plasmid pET-25b. Successful cloning was also confirmed by double enzymatic digestion, implying the presence of a sequence of 1250 bp as expected on 1% agarose gel [Figure 3]. After that, the sequence was sub-cloned between SalI and Ncol restriction sites in expression vector pET-32a (which has potential to increase the solubility of the recombinant protein).
Figure 3: Agarose gel electrophoresis 1% for cloning investigation. Cloning of polytopic construct in pET25b. Lane 1, upper band with ~5547 bp is intact vector and lower band showed with red arrow indicated a 1250 bp sequence (digestion of vector with BamH-I and Nco-I restriction enzymes); Lane 2, 10kb DNA ladder.Production of recombinant protein (expression and purification)
Multi-epitope sequence consisting of gp63, KMP- 11, HSP70, CPA, H2A, H3, as well as LACK and TRYP antigens from L. infantum was constructed by attaching them together by 4G1S flexible linkers. The construct was also successfully sub-cloned in pET-32a to be expressed in the prokaryotic system (E. coli BL21). SDS-PAGE and Western blotting of the bacterial cell lysate correspondingly demonstrated that the protein band was about 63kDa which agreed well with the calculated molecular weight according to [Figure 4]. Briefly, the purified recombinant protein and TRX-His-S-enterokinase fusion tag were analyzed on 12% SDS-PAGE. The prominent ~63 kDa and ~20 kDa bands corresponding to multi-epitope sequence and the tag were similarly observed on the SDS-PAGE, respectively. Furthermore, Western blotting approved the successful presence of these two bands [Figure 5] & [Figure 6].
Figure 4: SDS-PAGE and western blot analysis (Expression situation of MRP in E. coli harboring the construct. (A) SDS-PAGE 12%: Lane 1, molecular mass marker; Lane2, non-induced (pellete); lane 3, non-induced (supe) and lane 4 induced bacteria (supe). (B) Western blot: Lane 4, induced bacteria (supe).
Figure 5: Analysis of MRP after purification. (A) SDS-PAGE 12% and (B) western blot. Fractions were collected during different steps. Lane 1, molecular mass markers; Lane 2, purified MRP after elution with 500mM imidazole; Lane 3, wash fraction; Lane 4, flow-through; Lanes 5, Lysate.
Figure 6: Analysis of trx tag control after purification procedure. (A) SDS-PAGE 12% and (B) western blot. Fractions were collected during different steps. Lane 1, molecular mass markers; Lane 2, purified trx after elution with 500mM imidazole; Lane 3, flow-through; Lanes 4, Lysate.ELISA test
The test was well setup using 0.5 μg/ml of MRP, 1: 1000 serum concentration, and 1:10000 peroxidase-conjugated anti-IgG dilutions. Fortunately, there was no obvious reaction between the thioredoxin (TRX) tag and different sera during the ELISA optimization. Subsequently, the serodiagnosis was performed on all serum samples (VL+, VL-, healthy person, other parasitic disease). It should be noted that a well without any primary antibody was also considered as blank with optical density (OD) equal to 0.061. [Figure 7] shows the ROC resulting from the ELISA test. According to the curve and based on the optimum cutoff value (1.052), sensitivity, specificity, and positive predictive value (PPV) were estimated by 93.1%, 77.4% and 80.48%; respectively. MRP function was also evaluated by Western blotting using positive HVL and toxoplasma serums. Ultimately, the results revealed that the functional recombinant antigen could successfully react with positive VL sera although no reaction with toxoplasma serum sample was observed [Figure 8].
Figure 7: ROC curve analysis. The red line indicates the Youden index, providing a sensitivity of 93.1% and a specificity of 77.14% by optimum cut off value 1.052. The Area 0.94 indicates the overall ability of the test to discriminate between those individuals with VL and those without VL.
Figure 8: Western blot of MRP-IgG interaction. Lane 1, interaction between MRP and Toxo positive serum sample; Lane2, interaction between MRP and HVL positive serum sample; Lane 3, molecular mass marker and lane 4, MRP detected by anti poly-his peroxidase. Discussion Accurate and timely diagnosis is known as one of the measures essential for controlling leishmaniasis. Given that parasitological techniques for VL diagnosis are invasive, routine diagnosis of this disease can be based on serological methods. On the other hand, current methods for serological diagnosis of leishmaniasis do not always provide convincing results in different endemic areas worldwide. Maybe the main drawback of the situation is detectability of antileishmanial antibodies for a long period after recovery[14],[21]. Over the past two decades, attempts have been thus made to find an appropriate antigen for the resolution of this problem. One of the priorities in this area is the production of recombinant protein as an antigen for serological diagnosis[13].
In the past few decades, an increasing number of Leishmania antigens including most of the antigenic determinants of these protozoans have been consequently characterized. Eight antigenic markers derived from L. infantum in this study had been already tested as recombinant proteins, diagnostic markers, and immunogens in previous studies[28],[29],[30]. Due to the fact that each antigen alone may not have an effective immune interaction, a synthetic antigen made up of several different epitopes was likely to detect the immune response more strongly.
In recent years, results of different studies using multi-epitope antigens for CVL serodiagnosis have been successful[31],[32]. The only validated ELISA test using rK39 as an antigen has also revealed different results in various parts of the world (sensitivity: 88–100%; specificity: 86–100%)[33]. Moreover, a variety of diagnostic thresholds have been shown in different regions in Iran once using rK39 antigen for HVL serodiagnosis[34],[35],[36]. Recently, rK39 recombinant antigen from an Iranian strain of L. infantum has been evaluated by Hosseini Farash and collaborators to develop a local diagnostic test and to determine the suitability of this antigen in ELISA. They screened 84 human sera and 87 dog sera in endemic area of Meshkin-Shahr, Iran. The test was compared with the DAT for detection of anti- L. infantum antibodies and sensitivity/specificity were detected 85.7% and 86.0% respectively in human sera[18]. For example, rK28compared to rK39, as a new generation of fusion antigen was developed with an improved sensitivity (92-100%) attributable to low sensitivity of rk39 in Africa[37]. The result of another studyhave been similar in both of sensitivity and specificityby using rk28 and rk39 in the Indian subcontinent[38]. A meta-analysis by Maia concluded that investigators needed to consider use of both rK39 strip test and DAT before initiating anti- Leishmania treatment in endemic areas[39]. Furthermore, DAT as a routine serological test has been recommended for preliminary diagnosis of VL and seroepidemiological studies especially in endemic areas of Iran because the test is simple, cost effective and field applicable[40],[41]. A comparison between known VL serodiagnostic methods of DAT, ELISA test, and IFA in the city of Shiraz, Iran, also showed that DAT and ELISA test had the highest sensitivity and specificity[40]. Nevertheless, the sensitivity and specificity of each of these methods may vary in different endemic regions. So, the diagnostic tests selection could be based on the best sensitivity and specificity, the cost, the availability of equipment and qualified personnel, and field applicability[21]. The current pilot study aimed on promotion of the first parameter was developed with hope to use MRP in ELISA for detect HVL in endemic regions of Iran.
MRP Western blotting by HVL positive serums [Figure 8] showed two bands on the membrane (Lane 2) in which the upper band with a molecular weight of 63 kDa was the result of a reaction between HVL positive sera and recombinant protein with the expected size, but the lower pale band with a molecular weight range of 20 kDa might be related to the accidental truncated segment from the main recombinant protein. There was also no reaction between the positive toxoplasma serum and the recombinant protein in Lane 3. The results from the reaction of the anti-histidine antibody with the recombinant protein (Lane 4) stabilized that the 63 kDa band in Lane 2 was the same protein as the target one in the present study. The test revealed that a specific interaction was made between the antigen and anti-HVL antibody in the research. So, it was concluded that the MRP was endowed with enough capacity to be used as an antigen in the ELISA test. Sensitivity and specificity of the ELISA test using MRP were correspondingly reported by 93.1% and 77.4% in the present study; respectively. Maybe the main reasons for the decline in the specificity of the ELISA test were diversity of antigens utilized in this study, insufficient purity of the antigen and non-specific interaction with a small amount of residual bacterial protein in the elution.
Researchers have further shown that test specificity may increase especially in animal serum studies by reducing the diversity of antigens in the construct. For example, Soto reported a sensitivity of 79-93% and a specificity of 96-100% in CVL diagnosis using ribosomal proteins LP2a, LP2b, LiP0, and histone H2A[32]. A chimeric construct (K9, K26, and K39) against CVL and HVL also showed 99% specificity in both humans and canines, while sensitivity in canine was higher (96%) than human VL (82%)[31]. In general, these findings shed light on major challenges to the selection of immunodominant antigens to produce a mosaic form of the recombinant protein for VL diagnosis. Accordingly, indices such as selection and optimization of the antigenic determinants from L. infantum/chagasii; number, frequency, length, and immunological characteristics of the selected antigens, as well as attempts to express proteins in a eukaryotic host can be assumed crucial to produce a suitable recombinant. From another point of view, the sensitivity of chimeric antigens for HVL diagnosis was lower than for CVL diagnosis in several studies[18],[31],[42],[43]. Thus, it is possible that MRP will be more effective to CVL detection in the future investigations.
Based on the literature review, no study had been thus far published on the use of such a multi-epitope structure according to simultaneous use of Gp63, KMP-11, HSP70, CPA, H2A, H3, LACK and TRYP antigens for HVL diagnosis. It seems for determine final validity of the antigen in the endemic areas of VL in Iran, the TrxA tag must be separated from MRP for higher purity and more serum samples (positive, negative, and cross- reactive) will be need. Both of them require a different design to study in future.
Conflict of interest: None
References
留言 (0)